Loading...

XML

Word

Printable

Type: Epic
Resolution: Unresolved
Priority: Major
Fix Version/s: 2024Q3
Affects Version/s: None
Component/s: API, Data Pipeline
Labels:
- needs-product-input
- status_report

Epic Name:
Cost of AI loads (GPU)
Story Points:
2
Blocked:
False
Ready:
False
Epic Status:
To Do
Feature Link:
COST-5109 - Attribute the cost of running AI loads (GPU)
Parent Link:
COST-5109Attribute the cost of running AI loads (GPU)
Hierarchy Progress Bar:

33% To Do, 33% In Progress, 33% Done

SFDC Cases Links:
SFDC Cases Counter:
SFDC Cases Open:

PX Priority Data:
PX Impact Score:

Intelligence Requested:
Market:

Critical for OpenShift AI

Add GPU per-hour rate to price list in OCP cost models.
Support GPU dedicated
Support nvidia GPUs
On-premise, itemize cost of GPU, since it's a scarce resource (as compared to cost of CPU or memory or storage)

Notes

They would have to install the NVIDIA Operator to get the GPU metrics in Prometheus. See https://github.com/NVIDIA/dcgm-exporter.
https://docs.nvidia.com/datacenter/cloud-native/openshift/23.9.2/time-slicing-gpus-in-openshift.html
https://docs.openshift.com/container-platform/4.11/monitoring/nvidia-gpu-admin-dashboard.html

Epic Design Document

Feature Brainstorming Document

Kruize Research Documents

blocks

COST-3654 [Case 03473571] RFE: GPU Support in Cost Management

Closed

depends on

COST-5511 Release operator next with UBI 9, TSDB format, GPU/Namespace metrics

To Do

is blocked by

COST-5527 Add required GPU metrics to the operator reports for both COST and ROS

Backlog

is cloned by

COST-5060 Attribute the cost of running AI loads OCP on AWS (GPU)

Backlog

Assignee:: Unassigned

Reporter:: Kevan Holdaway

Votes:: 1 Vote for this issue

Watchers:: 20 Start watching this issue

Created:: 2023/04/26 8:46 PM

Updated:: 2024/11/18 4:26 PM